itsbenigno / ebg_stop

Multi-core EBGstop algorithm implementation

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Empirical Bernstein stopping

This is a C++ implementation of the Empirical Bernstein stopping algorithm (in particular EBGstop, a faster version of it).

Mnih, Volodymyr & Szepesvári, Csaba & Audibert, Jean-Yves. (2008). Empirical Bernstein stopping

Prerequisites

Make sure that your C++ compiler supports OpenMP

Run

N_THREADS defines how many parallel simulation of your random variable to run.

Modify the simulate_model() function with the random variable you want to estimate, then open a terminal inside the project directory and write

make

to compile and run the code.

About

Multi-core EBGstop algorithm implementation


Languages

Language:C++ 98.2%Language:Makefile 1.8%